Namespace(agnostic_nms=False, augment=False, batch_size=1, cfg='cfg/deploy/yolov7-flow.yaml', classes=None, conf_thres=0.25, device='', exist_ok=False, half_input=False, img_size=640, iou_thres=0.45, jit=False, mcore='MLU270', mlu_det=False, mname='offline-model', name='exp', no_trace=True, nosave=False, project='runs/detect', save=False, save_conf=False, save_txt=False, source='inference/format/training.jpg', update=False, view_img=False, weights=['mlu/weight/manymany.pth']) ==========Starting View Now============= cfg = cfg/deploy/yolov7-flow.yaml 2 cfg/deploy/yolov7-flow.yaml is yaml run on mlu ... the quantized model's weight: mlu/weight/manymany.pth ============= =============The model exactly after :============ ===layer_name:=== ['model.0.conv.weight', 'model.0.conv.scale', 'model.0.conv.quantized_mode', 'model.0.bn.weight', 'model.0.bn.bias', 'model.0.bn.running_mean', 'model.0.bn.running_var', 'model.0.bn.num_batches_tracked', 'model.1.conv.weight', 'model.1.conv.scale', 'model.1.conv.quantized_mode', 'model.1.bn.weight', 'model.1.bn.bias', 'model.1.bn.running_mean', 'model.1.bn.running_var', 'model.1.bn.num_batches_tracked', 'model.2.conv.weight', 'model.2.conv.scale', 'model.2.conv.quantized_mode', 'model.2.bn.weight', 'model.2.bn.bias', 'model.2.bn.running_mean', 'model.2.bn.running_var', 'model.2.bn.num_batches_tracked', 'model.3.conv.weight', 'model.3.conv.scale', 'model.3.conv.quantized_mode', 'model.3.bn.weight', 'model.3.bn.bias', 'model.3.bn.running_mean', 'model.3.bn.running_var', 'model.3.bn.num_batches_tracked', 'model.4.conv.weight', 'model.4.conv.scale', 'model.4.conv.quantized_mode', 'model.4.bn.weight', 'model.4.bn.bias', 'model.4.bn.running_mean', 'model.4.bn.running_var', 'model.4.bn.num_batches_tracked', 'model.5.conv.weight', 'model.5.conv.scale', 'model.5.conv.quantized_mode', 'model.5.bn.weight', 'model.5.bn.bias', 'model.5.bn.running_mean', 'model.5.bn.running_var', 'model.5.bn.num_batches_tracked', 'model.6.conv.weight', 'model.6.conv.scale', 'model.6.conv.quantized_mode', 'model.6.bn.weight', 'model.6.bn.bias', 'model.6.bn.running_mean', 'model.6.bn.running_var', 'model.6.bn.num_batches_tracked', 'model.7.conv.weight', 'model.7.conv.scale', 'model.7.conv.quantized_mode', 'model.7.bn.weight', 'model.7.bn.bias', 'model.7.bn.running_mean', 'model.7.bn.running_var', 'model.7.bn.num_batches_tracked', 'model.8.conv.weight', 'model.8.conv.scale', 'model.8.conv.quantized_mode', 'model.8.bn.weight', 'model.8.bn.bias', 'model.8.bn.running_mean', 'model.8.bn.running_var', 'model.8.bn.num_batches_tracked', 'model.9.conv.weight', 'model.9.conv.scale', 'model.9.conv.quantized_mode', 'model.9.bn.weight', 'model.9.bn.bias', 'model.9.bn.running_mean', 'model.9.bn.running_var', 'model.9.bn.num_batches_tracked', 'model.11.conv.weight', 'model.11.conv.scale', 'model.11.conv.quantized_mode', 'model.11.bn.weight', 'model.11.bn.bias', 'model.11.bn.running_mean', 'model.11.bn.running_var', 'model.11.bn.num_batches_tracked', 'model.13.conv.weight', 'model.13.conv.scale', 'model.13.conv.quantized_mode', 'model.13.bn.weight', 'model.13.bn.bias', 'model.13.bn.running_mean', 'model.13.bn.running_var', 'model.13.bn.num_batches_tracked', 'model.14.conv.weight', 'model.14.conv.scale', 'model.14.conv.quantized_mode', 'model.14.bn.weight', 'model.14.bn.bias', 'model.14.bn.running_mean', 'model.14.bn.running_var', 'model.14.bn.num_batches_tracked', 'model.15.conv.weight', 'model.15.conv.scale', 'model.15.conv.quantized_mode', 'model.15.bn.weight', 'model.15.bn.bias', 'model.15.bn.running_mean', 'model.15.bn.running_var', 'model.15.bn.num_batches_tracked', 'model.17.conv.weight', 'model.17.conv.scale', 'model.17.conv.quantized_mode', 'model.17.bn.weight', 'model.17.bn.bias', 'model.17.bn.running_mean', 'model.17.bn.running_var', 'model.17.bn.num_batches_tracked', 'model.18.conv.weight', 'model.18.conv.scale', 'model.18.conv.quantized_mode', 'model.18.bn.weight', 'model.18.bn.bias', 'model.18.bn.running_mean', 'model.18.bn.running_var', 'model.18.bn.num_batches_tracked', 'model.19.conv.weight', 'model.19.conv.scale', 'model.19.conv.quantized_mode', 'model.19.bn.weight', 'model.19.bn.bias', 'model.19.bn.running_mean', 'model.19.bn.running_var', 'model.19.bn.num_batches_tracked', 'model.20.conv.weight', 'model.20.conv.scale', 'model.20.conv.quantized_mode', 'model.20.bn.weight', 'model.20.bn.bias', 'model.20.bn.running_mean', 'model.20.bn.running_var', 'model.20.bn.num_batches_tracked', 'model.21.conv.weight', 'model.21.conv.scale', 'model.21.conv.quantized_mode', 'model.21.bn.weight', 'model.21.bn.bias', 'model.21.bn.running_mean', 'model.21.bn.running_var', 'model.21.bn.num_batches_tracked', 'model.22.conv.weight', 'model.22.conv.scale', 'model.22.conv.quantized_mode', 'model.22.bn.weight', 'model.22.bn.bias', 'model.22.bn.running_mean', 'model.22.bn.running_var', 'model.22.bn.num_batches_tracked', 'model.24.conv.weight', 'model.24.conv.scale', 'model.24.conv.quantized_mode', 'model.24.bn.weight', 'model.24.bn.bias', 'model.24.bn.running_mean', 'model.24.bn.running_var', 'model.24.bn.num_batches_tracked', 'model.26.conv.weight', 'model.26.conv.scale', 'model.26.conv.quantized_mode', 'model.26.bn.weight', 'model.26.bn.bias', 'model.26.bn.running_mean', 'model.26.bn.running_var', 'model.26.bn.num_batches_tracked', 'model.27.conv.weight', 'model.27.conv.scale', 'model.27.conv.quantized_mode', 'model.27.bn.weight', 'model.27.bn.bias', 'model.27.bn.running_mean', 'model.27.bn.running_var', 'model.27.bn.num_batches_tracked', 'model.28.conv.weight', 'model.28.conv.scale', 'model.28.conv.quantized_mode', 'model.28.bn.weight', 'model.28.bn.bias', 'model.28.bn.running_mean', 'model.28.bn.running_var', 'model.28.bn.num_batches_tracked', 'model.30.conv.weight', 'model.30.conv.scale', 'model.30.conv.quantized_mode', 'model.30.bn.weight', 'model.30.bn.bias', 'model.30.bn.running_mean', 'model.30.bn.running_var', 'model.30.bn.num_batches_tracked', 'model.31.conv.weight', 'model.31.conv.scale', 'model.31.conv.quantized_mode', 'model.31.bn.weight', 'model.31.bn.bias', 'model.31.bn.running_mean', 'model.31.bn.running_var', 'model.31.bn.num_batches_tracked', 'model.32.conv.weight', 'model.32.conv.scale', 'model.32.conv.quantized_mode', 'model.32.bn.weight', 'model.32.bn.bias', 'model.32.bn.running_mean', 'model.32.bn.running_var', 'model.32.bn.num_batches_tracked', 'model.33.conv.weight', 'model.33.conv.scale', 'model.33.conv.quantized_mode', 'model.33.bn.weight', 'model.33.bn.bias', 'model.33.bn.running_mean', 'model.33.bn.running_var', 'model.33.bn.num_batches_tracked', 'model.34.conv.weight', 'model.34.conv.scale', 'model.34.conv.quantized_mode', 'model.34.bn.weight', 'model.34.bn.bias', 'model.34.bn.running_mean', 'model.34.bn.running_var', 'model.34.bn.num_batches_tracked', 'model.35.conv.weight', 'model.35.conv.scale', 'model.35.conv.quantized_mode', 'model.35.bn.weight', 'model.35.bn.bias', 'model.35.bn.running_mean', 'model.35.bn.running_var', 'model.35.bn.num_batches_tracked', 'model.37.conv.weight', 'model.37.conv.scale', 'model.37.conv.quantized_mode', 'model.37.bn.weight', 'model.37.bn.bias', 'model.37.bn.running_mean', 'model.37.bn.running_var', 'model.37.bn.num_batches_tracked', 'model.39.conv.weight', 'model.39.conv.scale', 'model.39.conv.quantized_mode', 'model.39.bn.weight', 'model.39.bn.bias', 'model.39.bn.running_mean', 'model.39.bn.running_var', 'model.39.bn.num_batches_tracked', 'model.40.conv.weight', 'model.40.conv.scale', 'model.40.conv.quantized_mode', 'model.40.bn.weight', 'model.40.bn.bias', 'model.40.bn.running_mean', 'model.40.bn.running_var', 'model.40.bn.num_batches_tracked', 'model.41.conv.weight', 'model.41.conv.scale', 'model.41.conv.quantized_mode', 'model.41.bn.weight', 'model.41.bn.bias', 'model.41.bn.running_mean', 'model.41.bn.running_var', 'model.41.bn.num_batches_tracked', 'model.43.conv.weight', 'model.43.conv.scale', 'model.43.conv.quantized_mode', 'model.43.bn.weight', 'model.43.bn.bias', 'model.43.bn.running_mean', 'model.43.bn.running_var', 'model.43.bn.num_batches_tracked', 'model.44.conv.weight', 'model.44.conv.scale', 'model.44.conv.quantized_mode', 'model.44.bn.weight', 'model.44.bn.bias', 'model.44.bn.running_mean', 'model.44.bn.running_var', 'model.44.bn.num_batches_tracked', 'model.45.conv.weight', 'model.45.conv.scale', 'model.45.conv.quantized_mode', 'model.45.bn.weight', 'model.45.bn.bias', 'model.45.bn.running_mean', 'model.45.bn.running_var', 'model.45.bn.num_batches_tracked', 'model.46.conv.weight', 'model.46.conv.scale', 'model.46.conv.quantized_mode', 'model.46.bn.weight', 'model.46.bn.bias', 'model.46.bn.running_mean', 'model.46.bn.running_var', 'model.46.bn.num_batches_tracked', 'model.47.conv.weight', 'model.47.conv.scale', 'model.47.conv.quantized_mode', 'model.47.bn.weight', 'model.47.bn.bias', 'model.47.bn.running_mean', 'model.47.bn.running_var', 'model.47.bn.num_batches_tracked', 'model.48.conv.weight', 'model.48.conv.scale', 'model.48.conv.quantized_mode', 'model.48.bn.weight', 'model.48.bn.bias', 'model.48.bn.running_mean', 'model.48.bn.running_var', 'model.48.bn.num_batches_tracked', 'model.50.conv.weight', 'model.50.conv.scale', 'model.50.conv.quantized_mode', 'model.50.bn.weight', 'model.50.bn.bias', 'model.50.bn.running_mean', 'model.50.bn.running_var', 'model.50.bn.num_batches_tracked', 'model.51.cv1.conv.weight', 'model.51.cv1.conv.scale', 'model.51.cv1.conv.quantized_mode', 'model.51.cv1.bn.weight', 'model.51.cv1.bn.bias', 'model.51.cv1.bn.running_mean', 'model.51.cv1.bn.running_var', 'model.51.cv1.bn.num_batches_tracked', 'model.51.cv2.conv.weight', 'model.51.cv2.conv.scale', 'model.51.cv2.conv.quantized_mode', 'model.51.cv2.bn.weight', 'model.51.cv2.bn.bias', 'model.51.cv2.bn.running_mean', 'model.51.cv2.bn.running_var', 'model.51.cv2.bn.num_batches_tracked', 'model.51.cv3.conv.weight', 'model.51.cv3.conv.scale', 'model.51.cv3.conv.quantized_mode', 'model.51.cv3.bn.weight', 'model.51.cv3.bn.bias', 'model.51.cv3.bn.running_mean', 'model.51.cv3.bn.running_var', 'model.51.cv3.bn.num_batches_tracked', 'model.51.cv4.conv.weight', 'model.51.cv4.conv.scale', 'model.51.cv4.conv.quantized_mode', 'model.51.cv4.bn.weight', 'model.51.cv4.bn.bias', 'model.51.cv4.bn.running_mean', 'model.51.cv4.bn.running_var', 'model.51.cv4.bn.num_batches_tracked', 'model.51.cv5.conv.weight', 'model.51.cv5.conv.scale', 'model.51.cv5.conv.quantized_mode', 'model.51.cv5.bn.weight', 'model.51.cv5.bn.bias', 'model.51.cv5.bn.running_mean', 'model.51.cv5.bn.running_var', 'model.51.cv5.bn.num_batches_tracked', 'model.51.cv6.conv.weight', 'model.51.cv6.conv.scale', 'model.51.cv6.conv.quantized_mode', 'model.51.cv6.bn.weight', 'model.51.cv6.bn.bias', 'model.51.cv6.bn.running_mean', 'model.51.cv6.bn.running_var', 'model.51.cv6.bn.num_batches_tracked', 'model.51.cv7.conv.weight', 'model.51.cv7.conv.scale', 'model.51.cv7.conv.quantized_mode', 'model.51.cv7.bn.weight', 'model.51.cv7.bn.bias', 'model.51.cv7.bn.running_mean', 'model.51.cv7.bn.running_var', 'model.51.cv7.bn.num_batches_tracked', 'model.52.conv.weight', 'model.52.conv.scale', 'model.52.conv.quantized_mode', 'model.52.bn.weight', 'model.52.bn.bias', 'model.52.bn.running_mean', 'model.52.bn.running_var', 'model.52.bn.num_batches_tracked', 'model.54.conv.weight', 'model.54.conv.scale', 'model.54.conv.quantized_mode', 'model.54.bn.weight', 'model.54.bn.bias', 'model.54.bn.running_mean', 'model.54.bn.running_var', 'model.54.bn.num_batches_tracked', 'model.56.conv.weight', 'model.56.conv.scale', 'model.56.conv.quantized_mode', 'model.56.bn.weight', 'model.56.bn.bias', 'model.56.bn.running_mean', 'model.56.bn.running_var', 'model.56.bn.num_batches_tracked', 'model.57.conv.weight', 'model.57.conv.scale', 'model.57.conv.quantized_mode', 'model.57.bn.weight', 'model.57.bn.bias', 'model.57.bn.running_mean', 'model.57.bn.running_var', 'model.57.bn.num_batches_tracked', 'model.58.conv.weight', 'model.58.conv.scale', 'model.58.conv.quantized_mode', 'model.58.bn.weight', 'model.58.bn.bias', 'model.58.bn.running_mean', 'model.58.bn.running_var', 'model.58.bn.num_batches_tracked', 'model.59.conv.weight', 'model.59.conv.scale', 'model.59.conv.quantized_mode', 'model.59.bn.weight', 'model.59.bn.bias', 'model.59.bn.running_mean', 'model.59.bn.running_var', 'model.59.bn.num_batches_tracked', 'model.60.conv.weight', 'model.60.conv.scale', 'model.60.conv.quantized_mode', 'model.60.bn.weight', 'model.60.bn.bias', 'model.60.bn.running_mean', 'model.60.bn.running_var', 'model.60.bn.num_batches_tracked', 'model.61.conv.weight', 'model.61.conv.scale', 'model.61.conv.quantized_mode', 'model.61.bn.weight', 'model.61.bn.bias', 'model.61.bn.running_mean', 'model.61.bn.running_var', 'model.61.bn.num_batches_tracked', 'model.63.conv.weight', 'model.63.conv.scale', 'model.63.conv.quantized_mode', 'model.63.bn.weight', 'model.63.bn.bias', 'model.63.bn.running_mean', 'model.63.bn.running_var', 'model.63.bn.num_batches_tracked', 'model.64.conv.weight', 'model.64.conv.scale', 'model.64.conv.quantized_mode', 'model.64.bn.weight', 'model.64.bn.bias', 'model.64.bn.running_mean', 'model.64.bn.running_var', 'model.64.bn.num_batches_tracked', 'model.66.conv.weight', 'model.66.conv.scale', 'model.66.conv.quantized_mode', 'model.66.bn.weight', 'model.66.bn.bias', 'model.66.bn.running_mean', 'model.66.bn.running_var', 'model.66.bn.num_batches_tracked', 'model.68.conv.weight', 'model.68.conv.scale', 'model.68.conv.quantized_mode', 'model.68.bn.weight', 'model.68.bn.bias', 'model.68.bn.running_mean', 'model.68.bn.running_var', 'model.68.bn.num_batches_tracked', 'model.69.conv.weight', 'model.69.conv.scale', 'model.69.conv.quantized_mode', 'model.69.bn.weight', 'model.69.bn.bias', 'model.69.bn.running_mean', 'model.69.bn.running_var', 'model.69.bn.num_batches_tracked', 'model.70.conv.weight', 'model.70.conv.scale', 'model.70.conv.quantized_mode', 'model.70.bn.weight', 'model.70.bn.bias', 'model.70.bn.running_mean', 'model.70.bn.running_var', 'model.70.bn.num_batches_tracked', 'model.71.conv.weight', 'model.71.conv.scale', 'model.71.conv.quantized_mode', 'model.71.bn.weight', 'model.71.bn.bias', 'model.71.bn.running_mean', 'model.71.bn.running_var', 'model.71.bn.num_batches_tracked', 'model.72.conv.weight', 'model.72.conv.scale', 'model.72.conv.quantized_mode', 'model.72.bn.weight', 'model.72.bn.bias', 'model.72.bn.running_mean', 'model.72.bn.running_var', 'model.72.bn.num_batches_tracked', 'model.73.conv.weight', 'model.73.conv.scale', 'model.73.conv.quantized_mode', 'model.73.bn.weight', 'model.73.bn.bias', 'model.73.bn.running_mean', 'model.73.bn.running_var', 'model.73.bn.num_batches_tracked', 'model.75.conv.weight', 'model.75.conv.scale', 'model.75.conv.quantized_mode', 'model.75.bn.weight', 'model.75.bn.bias', 'model.75.bn.running_mean', 'model.75.bn.running_var', 'model.75.bn.num_batches_tracked', 'model.77.conv.weight', 'model.77.conv.scale', 'model.77.conv.quantized_mode', 'model.77.bn.weight', 'model.77.bn.bias', 'model.77.bn.running_mean', 'model.77.bn.running_var', 'model.77.bn.num_batches_tracked', 'model.78.conv.weight', 'model.78.conv.scale', 'model.78.conv.quantized_mode', 'model.78.bn.weight', 'model.78.bn.bias', 'model.78.bn.running_mean', 'model.78.bn.running_var', 'model.78.bn.num_batches_tracked', 'model.79.conv.weight', 'model.79.conv.scale', 'model.79.conv.quantized_mode', 'model.79.bn.weight', 'model.79.bn.bias', 'model.79.bn.running_mean', 'model.79.bn.running_var', 'model.79.bn.num_batches_tracked', 'model.81.conv.weight', 'model.81.conv.scale', 'model.81.conv.quantized_mode', 'model.81.bn.weight', 'model.81.bn.bias', 'model.81.bn.running_mean', 'model.81.bn.running_var', 'model.81.bn.num_batches_tracked', 'model.82.conv.weight', 'model.82.conv.scale', 'model.82.conv.quantized_mode', 'model.82.bn.weight', 'model.82.bn.bias', 'model.82.bn.running_mean', 'model.82.bn.running_var', 'model.82.bn.num_batches_tracked', 'model.83.conv.weight', 'model.83.conv.scale', 'model.83.conv.quantized_mode', 'model.83.bn.weight', 'model.83.bn.bias', 'model.83.bn.running_mean', 'model.83.bn.running_var', 'model.83.bn.num_batches_tracked', 'model.84.conv.weight', 'model.84.conv.scale', 'model.84.conv.quantized_mode', 'model.84.bn.weight', 'model.84.bn.bias', 'model.84.bn.running_mean', 'model.84.bn.running_var', 'model.84.bn.num_batches_tracked', 'model.85.conv.weight', 'model.85.conv.scale', 'model.85.conv.quantized_mode', 'model.85.bn.weight', 'model.85.bn.bias', 'model.85.bn.running_mean', 'model.85.bn.running_var', 'model.85.bn.num_batches_tracked', 'model.86.conv.weight', 'model.86.conv.scale', 'model.86.conv.quantized_mode', 'model.86.bn.weight', 'model.86.bn.bias', 'model.86.bn.running_mean', 'model.86.bn.running_var', 'model.86.bn.num_batches_tracked', 'model.88.conv.weight', 'model.88.conv.scale', 'model.88.conv.quantized_mode', 'model.88.bn.weight', 'model.88.bn.bias', 'model.88.bn.running_mean', 'model.88.bn.running_var', 'model.88.bn.num_batches_tracked', 'model.90.conv.weight', 'model.90.conv.scale', 'model.90.conv.quantized_mode', 'model.90.bn.weight', 'model.90.bn.bias', 'model.90.bn.running_mean', 'model.90.bn.running_var', 'model.90.bn.num_batches_tracked', 'model.91.conv.weight', 'model.91.conv.scale', 'model.91.conv.quantized_mode', 'model.91.bn.weight', 'model.91.bn.bias', 'model.91.bn.running_mean', 'model.91.bn.running_var', 'model.91.bn.num_batches_tracked', 'model.92.conv.weight', 'model.92.conv.scale', 'model.92.conv.quantized_mode', 'model.92.bn.weight', 'model.92.bn.bias', 'model.92.bn.running_mean', 'model.92.bn.running_var', 'model.92.bn.num_batches_tracked', 'model.94.conv.weight', 'model.94.conv.scale', 'model.94.conv.quantized_mode', 'model.94.bn.weight', 'model.94.bn.bias', 'model.94.bn.running_mean', 'model.94.bn.running_var', 'model.94.bn.num_batches_tracked', 'model.95.conv.weight', 'model.95.conv.scale', 'model.95.conv.quantized_mode', 'model.95.bn.weight', 'model.95.bn.bias', 'model.95.bn.running_mean', 'model.95.bn.running_var', 'model.95.bn.num_batches_tracked', 'model.96.conv.weight', 'model.96.conv.scale', 'model.96.conv.quantized_mode', 'model.96.bn.weight', 'model.96.bn.bias', 'model.96.bn.running_mean', 'model.96.bn.running_var', 'model.96.bn.num_batches_tracked', 'model.97.conv.weight', 'model.97.conv.scale', 'model.97.conv.quantized_mode', 'model.97.bn.weight', 'model.97.bn.bias', 'model.97.bn.running_mean', 'model.97.bn.running_var', 'model.97.bn.num_batches_tracked', 'model.98.conv.weight', 'model.98.conv.scale', 'model.98.conv.quantized_mode', 'model.98.bn.weight', 'model.98.bn.bias', 'model.98.bn.running_mean', 'model.98.bn.running_var', 'model.98.bn.num_batches_tracked', 'model.99.conv.weight', 'model.99.conv.scale', 'model.99.conv.quantized_mode', 'model.99.bn.weight', 'model.99.bn.bias', 'model.99.bn.running_mean', 'model.99.bn.running_var', 'model.99.bn.num_batches_tracked', 'model.101.conv.weight', 'model.101.conv.scale', 'model.101.conv.quantized_mode', 'model.101.bn.weight', 'model.101.bn.bias', 'model.101.bn.running_mean', 'model.101.bn.running_var', 'model.101.bn.num_batches_tracked', 'model.102.rbr_dense.0.weight', 'model.102.rbr_dense.0.scale', 'model.102.rbr_dense.0.quantized_mode', 'model.102.rbr_dense.1.weight', 'model.102.rbr_dense.1.bias', 'model.102.rbr_dense.1.running_mean', 'model.102.rbr_dense.1.running_var', 'model.102.rbr_dense.1.num_batches_tracked', 'model.102.rbr_1x1.0.weight', 'model.102.rbr_1x1.0.scale', 'model.102.rbr_1x1.0.quantized_mode', 'model.102.rbr_1x1.1.weight', 'model.102.rbr_1x1.1.bias', 'model.102.rbr_1x1.1.running_mean', 'model.102.rbr_1x1.1.running_var', 'model.102.rbr_1x1.1.num_batches_tracked', 'model.103.rbr_dense.0.weight', 'model.103.rbr_dense.0.scale', 'model.103.rbr_dense.0.quantized_mode', 'model.103.rbr_dense.1.weight', 'model.103.rbr_dense.1.bias', 'model.103.rbr_dense.1.running_mean', 'model.103.rbr_dense.1.running_var', 'model.103.rbr_dense.1.num_batches_tracked', 'model.103.rbr_1x1.0.weight', 'model.103.rbr_1x1.0.scale', 'model.103.rbr_1x1.0.quantized_mode', 'model.103.rbr_1x1.1.weight', 'model.103.rbr_1x1.1.bias', 'model.103.rbr_1x1.1.running_mean', 'model.103.rbr_1x1.1.running_var', 'model.103.rbr_1x1.1.num_batches_tracked', 'model.104.rbr_dense.0.weight', 'model.104.rbr_dense.0.scale', 'model.104.rbr_dense.0.quantized_mode', 'model.104.rbr_dense.1.weight', 'model.104.rbr_dense.1.bias', 'model.104.rbr_dense.1.running_mean', 'model.104.rbr_dense.1.running_var', 'model.104.rbr_dense.1.num_batches_tracked', 'model.104.rbr_1x1.0.weight', 'model.104.rbr_1x1.0.scale', 'model.104.rbr_1x1.0.quantized_mode', 'model.104.rbr_1x1.1.weight', 'model.104.rbr_1x1.1.bias', 'model.104.rbr_1x1.1.running_mean', 'model.104.rbr_1x1.1.running_var', 'model.104.rbr_1x1.1.num_batches_tracked', 'model.105.anchors', 'model.105.anchor_grid', 'model.105.m.0.weight', 'model.105.m.0.bias', 'model.105.m.0.scale', 'model.105.m.0.quantized_mode', 'model.105.m.1.weight', 'model.105.m.1.bias', 'model.105.m.1.scale', 'model.105.m.1.quantized_mode', 'model.105.m.2.weight', 'model.105.m.2.bias', 'model.105.m.2.scale', 'model.105.m.2.quantized_mode', 'model.105.ia.0.implicit', 'model.105.ia.1.implicit', 'model.105.ia.2.implicit', 'model.105.im.0.implicit', 'model.105.im.1.implicit', 'model.105.im.2.implicit'] ======The whole model printing:====== Model( (model): Sequential( (0): Conv( (conv): MLUConv2d( 3, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([127.00000, 159.96063]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (1): Conv( (conv): MLUConv2d( 32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 1.88034, 242.17505]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (2): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 0.76717, 246.77040]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (3): Conv( (conv): MLUConv2d( 64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 2.16945, 659.72351]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (4): Conv( (conv): MLUConv2d( 128, 64, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.92369, 628.25122]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (5): Conv( (conv): MLUConv2d( 128, 64, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.92369, 475.93048]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (6): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 6.41040, 554.28021]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (7): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.03297, 458.72311]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (8): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.11026, 767.81110]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (9): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 9.95951, 859.82147]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (10): Concat() (11): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.41040, 273.06668]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (12): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (13): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 8.11695, 501.63162]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (14): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 8.11695, 685.81677]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (15): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 6.14557, 988.01898]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (16): Concat() (17): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 7.33305, 321.70190]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (18): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 7.33305, 435.67169]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (19): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 8.65670, 636.32050]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (20): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 9.67595, 407.03598]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (21): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 9.30384, 817.26947]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (22): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 8.85014, 893.03345]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (23): Concat() (24): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 5.68211, 132.76978]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (25): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (26): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.48813, 743.66260]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (27): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.48813, 1697.20068]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (28): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 6.09056, 2152.88989]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (29): Concat() (30): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 5.49991, 1761.86963]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (31): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 5.49991, 1102.68579]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (32): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 4.42657, 1578.73145]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (33): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.95019, 1872.87854]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (34): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.66529, 2285.30249]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (35): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.20015, 2676.22900]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (36): Concat() (37): Conv( (conv): MLUConv2d( 1024, 1024, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.37654, 1072.00830]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (38): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (39): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.97813, 1084.86340]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (40): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.97813, 1796.86353]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (41): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([2.79315e+00, 3.12897e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (42): Concat() (43): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.87262, 2285.30249]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (44): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.87262, 1101.51831]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (45): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.47857, 2038.96912]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (46): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.55337, 1933.79932]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (47): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 1.96922, 1937.40039]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (48): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.46558, 2126.48755]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (49): Concat() (50): Conv( (conv): MLUConv2d( 1024, 1024, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([1.96922e+00, 2.21594e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (51): SPPCSPC( (cv1): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.68248, 2660.82861]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([2.68248e+00, 2.89196e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.46504, 1826.83752]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv4): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.22379, 2116.75269]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): ModuleList( (0): MaxPool2d(kernel_size=5, stride=1, padding=2, dilation=1, ceil_mode=False) (1): MaxPool2d(kernel_size=9, stride=1, padding=4, dilation=1, ceil_mode=False) (2): MaxPool2d(kernel_size=13, stride=1, padding=6, dilation=1, ceil_mode=False) ) (cv5): Conv( (conv): MLUConv2d( 2048, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.67165, 1078.67700]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv6): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 6.95111, 1847.92896]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv7): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.59052, 1191.73425]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (52): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.05947, 1121.70776]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (53): Upsample(scale_factor=2.0, mode=nearest) (54): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.97813, 747.93964]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (55): Concat() (56): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.76734, 580.24762]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (57): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.76734, 463.21637]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (58): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.38502, 1523.25623]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (59): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.50332, 1491.58997]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (60): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.87678, 2361.82520]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (61): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.08313, 2051.02808]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (62): Concat() (63): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.09477, 108.37334]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (64): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.04031, 195.12077]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (65): Upsample(scale_factor=2.0, mode=nearest) (66): Conv( (conv): MLUConv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.48813, 98.59591]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (67): Concat() (68): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.61115, 108.46372]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (69): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.61115, 48.56161]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (70): Conv( (conv): MLUConv2d( 128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.58413, 58.05714]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (71): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 4.80973, 54.36789]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (72): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.31641, 159.56810]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (73): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.03452, 221.17007]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (74): Concat() (75): Conv( (conv): MLUConv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.58413, 81.63716]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (76): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (77): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 4.96912, 231.19644]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (78): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 4.96912, 157.72954]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (79): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 4.73306, 803.38531]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (80): Concat() (81): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.04031, 527.04358]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (82): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.04031, 605.22632]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (83): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.66400, 776.98584]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (84): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.54726, 1158.55676]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (85): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.66257, 892.26758]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (86): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.70854, 1283.63232]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (87): Concat() (88): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.63626, 660.56128]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (89): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (90): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([1.42696e+00, 1.93921e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (91): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.42696, 1184.27319]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (92): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([2.26463e+00, 2.66936e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (93): Concat() (94): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([2.05947e+00, 3.00255e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (95): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.05947, 2010.40381]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (96): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.60792, 2324.88037]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (97): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.54627, 2320.99048]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (98): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.66854, 1679.39307]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (99): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.67804, 2186.82910]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (100): Concat() (101): Conv( (conv): MLUConv2d( 2048, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.54117, 1505.62085]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (102): RepConv( (act): SiLU() (rbr_dense): Sequential( (0): MLUConv2d( 128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 4.96912, 691.28503]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) (rbr_1x1): Sequential( (0): MLUConv2d( 128, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 4.96912, 801.52850]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) ) (103): RepConv( (act): SiLU() (rbr_dense): Sequential( (0): MLUConv2d( 256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 1.42696, 587.12415]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) (rbr_1x1): Sequential( (0): MLUConv2d( 256, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.42696, 419.50967]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) ) (104): RepConv( (act): SiLU() (rbr_dense): Sequential( (0): MLUConv2d( 512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([1.79659e+00, 2.11460e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) (rbr_1x1): Sequential( (0): MLUConv2d( 512, 1024, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.79659, 1612.99841]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) ) (105): IDetect( (m): ModuleList( (0): MLUConv2d( 256, 18, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.20508, 377.77197]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): MLUConv2d( 512, 18, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.24337, 245.83743]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (2): MLUConv2d( 1024, 18, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.27830, 1221.82507]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) ) (ia): ModuleList( (0): ImplicitA() (1): ImplicitA() (2): ImplicitA() ) (im): ModuleList( (0): ImplicitM() (1): ImplicitM() (2): ImplicitM() ) ) ) ) ======The whole model torchinfo:====== ？？？？？？？？？？？？为何一开始就是SPPCSPC+IDetect？还是说这是torchinfo的问题，不能反映量化问题？？？？？？ ========================================================================================== Layer (type:depth-idx) Output Shape Param # ========================================================================================== Model -- -- ├─Sequential: 1 -- -- │ └─SPPCSPC: 2 -- -- │ │ └─ModuleList: 3-1 -- -- │ └─IDetect: 2 -- -- │ │ └─ModuleList: 3-2 -- 32,319 │ │ └─ModuleList: 3-3 -- 1,792 │ │ └─ModuleList: 3-4 -- 54 │ └─Conv: 2-1 [1, 32, 640, 640] -- │ │ └─MLUConv2d: 3-5 [1, 32, 640, 640] 899 │ │ └─BatchNorm2d: 3-6 [1, 32, 640, 640] 64 │ │ └─SiLU: 3-7 [1, 32, 640, 640] -- │ └─Conv: 2-2 [1, 64, 320, 320] -- │ │ └─MLUConv2d: 3-8 [1, 64, 320, 320] 18,499 │ │ └─BatchNorm2d: 3-9 [1, 64, 320, 320] 128 │ │ └─SiLU: 3-10 [1, 64, 320, 320] -- │ └─Conv: 2-3 [1, 64, 320, 320] -- │ │ └─MLUConv2d: 3-11 [1, 64, 320, 320] 36,931 │ │ └─BatchNorm2d: 3-12 [1, 64, 320, 320] 128 │ │ └─SiLU: 3-13 [1, 64, 320, 320] -- │ └─Conv: 2-4 [1, 128, 160, 160] -- │ │ └─MLUConv2d: 3-14 [1, 128, 160, 160] 73,859 │ │ └─BatchNorm2d: 3-15 [1, 128, 160, 160] 256 │ │ └─SiLU: 3-16 [1, 128, 160, 160] -- │ └─Conv: 2-5 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-17 [1, 64, 160, 160] 8,259 │ │ └─BatchNorm2d: 3-18 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-19 [1, 64, 160, 160] -- │ └─Conv: 2-6 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-20 [1, 64, 160, 160] 8,259 │ │ └─BatchNorm2d: 3-21 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-22 [1, 64, 160, 160] -- │ └─Conv: 2-7 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-23 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-24 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-25 [1, 64, 160, 160] -- │ └─Conv: 2-8 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-26 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-27 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-28 [1, 64, 160, 160] -- │ └─Conv: 2-9 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-29 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-30 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-31 [1, 64, 160, 160] -- │ └─Conv: 2-10 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-32 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-33 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-34 [1, 64, 160, 160] -- │ └─Concat: 2-11 [1, 256, 160, 160] -- │ └─Conv: 2-12 [1, 256, 160, 160] -- │ │ └─MLUConv2d: 3-35 [1, 256, 160, 160] 65,795 │ │ └─BatchNorm2d: 3-36 [1, 256, 160, 160] 512 │ │ └─SiLU: 3-37 [1, 256, 160, 160] -- │ └─MP: 2-13 [1, 256, 80, 80] -- │ │ └─MaxPool2d: 3-38 [1, 256, 80, 80] -- │ └─Conv: 2-14 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-39 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-40 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-41 [1, 128, 80, 80] -- │ └─Conv: 2-15 [1, 128, 160, 160] -- │ │ └─MLUConv2d: 3-42 [1, 128, 160, 160] 32,899 │ │ └─BatchNorm2d: 3-43 [1, 128, 160, 160] 256 │ │ └─SiLU: 3-44 [1, 128, 160, 160] -- │ └─Conv: 2-16 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-45 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-46 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-47 [1, 128, 80, 80] -- │ └─Concat: 2-17 [1, 256, 80, 80] -- │ └─Conv: 2-18 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-48 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-49 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-50 [1, 128, 80, 80] -- │ └─Conv: 2-19 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-51 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-52 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-53 [1, 128, 80, 80] -- │ └─Conv: 2-20 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-54 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-55 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-56 [1, 128, 80, 80] -- │ └─Conv: 2-21 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-57 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-58 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-59 [1, 128, 80, 80] -- │ └─Conv: 2-22 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-60 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-61 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-62 [1, 128, 80, 80] -- │ └─Conv: 2-23 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-63 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-64 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-65 [1, 128, 80, 80] -- │ └─Concat: 2-24 [1, 512, 80, 80] -- │ └─Conv: 2-25 [1, 512, 80, 80] -- │ │ └─MLUConv2d: 3-66 [1, 512, 80, 80] 262,659 │ │ └─BatchNorm2d: 3-67 [1, 512, 80, 80] 1,024 │ │ └─SiLU: 3-68 [1, 512, 80, 80] -- │ └─MP: 2-26 [1, 512, 40, 40] -- │ │ └─MaxPool2d: 3-69 [1, 512, 40, 40] -- │ └─Conv: 2-27 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-70 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-71 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-72 [1, 256, 40, 40] -- │ └─Conv: 2-28 [1, 256, 80, 80] -- │ │ └─MLUConv2d: 3-73 [1, 256, 80, 80] 131,331 │ │ └─BatchNorm2d: 3-74 [1, 256, 80, 80] 512 │ │ └─SiLU: 3-75 [1, 256, 80, 80] -- │ └─Conv: 2-29 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-76 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-77 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-78 [1, 256, 40, 40] -- │ └─Concat: 2-30 [1, 512, 40, 40] -- │ └─Conv: 2-31 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-79 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-80 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-81 [1, 256, 40, 40] -- │ └─Conv: 2-32 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-82 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-83 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-84 [1, 256, 40, 40] -- │ └─Conv: 2-33 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-85 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-86 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-87 [1, 256, 40, 40] -- │ └─Conv: 2-34 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-88 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-89 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-90 [1, 256, 40, 40] -- │ └─Conv: 2-35 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-91 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-92 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-93 [1, 256, 40, 40] -- │ └─Conv: 2-36 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-94 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-95 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-96 [1, 256, 40, 40] -- │ └─Concat: 2-37 [1, 1024, 40, 40] -- │ └─Conv: 2-38 [1, 1024, 40, 40] -- │ │ └─MLUConv2d: 3-97 [1, 1024, 40, 40] 1,049,603 │ │ └─BatchNorm2d: 3-98 [1, 1024, 40, 40] 2,048 │ │ └─SiLU: 3-99 [1, 1024, 40, 40] -- │ └─MP: 2-39 [1, 1024, 20, 20] -- │ │ └─MaxPool2d: 3-100 [1, 1024, 20, 20] -- │ └─Conv: 2-40 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-101 [1, 512, 20, 20] 524,803 │ │ └─BatchNorm2d: 3-102 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-103 [1, 512, 20, 20] -- │ └─Conv: 2-41 [1, 512, 40, 40] -- │ │ └─MLUConv2d: 3-104 [1, 512, 40, 40] 524,803 │ │ └─BatchNorm2d: 3-105 [1, 512, 40, 40] 1,024 │ │ └─SiLU: 3-106 [1, 512, 40, 40] -- │ └─Conv: 2-42 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-107 [1, 512, 20, 20] 2,359,811 │ │ └─BatchNorm2d: 3-108 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-109 [1, 512, 20, 20] -- │ └─Concat: 2-43 [1, 1024, 20, 20] -- │ └─Conv: 2-44 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-110 [1, 256, 20, 20] 262,403 │ │ └─BatchNorm2d: 3-111 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-112 [1, 256, 20, 20] -- │ └─Conv: 2-45 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-113 [1, 256, 20, 20] 262,403 │ │ └─BatchNorm2d: 3-114 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-115 [1, 256, 20, 20] -- │ └─Conv: 2-46 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-116 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-117 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-118 [1, 256, 20, 20] -- │ └─Conv: 2-47 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-119 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-120 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-121 [1, 256, 20, 20] -- │ └─Conv: 2-48 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-122 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-123 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-124 [1, 256, 20, 20] -- │ └─Conv: 2-49 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-125 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-126 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-127 [1, 256, 20, 20] -- │ └─Concat: 2-50 [1, 1024, 20, 20] -- │ └─Conv: 2-51 [1, 1024, 20, 20] -- │ │ └─MLUConv2d: 3-128 [1, 1024, 20, 20] 1,049,603 │ │ └─BatchNorm2d: 3-129 [1, 1024, 20, 20] 2,048 │ │ └─SiLU: 3-130 [1, 1024, 20, 20] -- │ └─SPPCSPC: 2-52 [1, 512, 20, 20] -- │ │ └─Conv: 3-131 [1, 512, 20, 20] 525,827 │ │ └─Conv: 3-132 [1, 512, 20, 20] 2,360,835 │ │ └─Conv: 3-133 [1, 512, 20, 20] 263,683 │ │ └─Conv: 3-134 [1, 512, 20, 20] 1,050,115 │ │ └─Conv: 3-135 [1, 512, 20, 20] 2,360,835 │ │ └─Conv: 3-136 [1, 512, 20, 20] 525,827 │ │ └─Conv: 3-137 [1, 512, 20, 20] 525,827 │ └─Conv: 2-53 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-138 [1, 256, 20, 20] 131,331 │ │ └─BatchNorm2d: 3-139 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-140 [1, 256, 20, 20] -- │ └─Upsample: 2-54 [1, 256, 40, 40] -- │ └─Conv: 2-55 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-141 [1, 256, 40, 40] 262,403 │ │ └─BatchNorm2d: 3-142 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-143 [1, 256, 40, 40] -- │ └─Concat: 2-56 [1, 512, 40, 40] -- │ └─Conv: 2-57 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-144 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-145 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-146 [1, 256, 40, 40] -- │ └─Conv: 2-58 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-147 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-148 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-149 [1, 256, 40, 40] -- │ └─Conv: 2-59 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-150 [1, 128, 40, 40] 295,043 │ │ └─BatchNorm2d: 3-151 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-152 [1, 128, 40, 40] -- │ └─Conv: 2-60 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-153 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-154 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-155 [1, 128, 40, 40] -- │ └─Conv: 2-61 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-156 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-157 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-158 [1, 128, 40, 40] -- │ └─Conv: 2-62 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-159 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-160 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-161 [1, 128, 40, 40] -- │ └─Concat: 2-63 [1, 1024, 40, 40] -- │ └─Conv: 2-64 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-162 [1, 256, 40, 40] 262,403 │ │ └─BatchNorm2d: 3-163 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-164 [1, 256, 40, 40] -- │ └─Conv: 2-65 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-165 [1, 128, 40, 40] 32,899 │ │ └─BatchNorm2d: 3-166 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-167 [1, 128, 40, 40] -- │ └─Upsample: 2-66 [1, 128, 80, 80] -- │ └─Conv: 2-67 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-168 [1, 128, 80, 80] 65,667 │ │ └─BatchNorm2d: 3-169 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-170 [1, 128, 80, 80] -- │ └─Concat: 2-68 [1, 256, 80, 80] -- │ └─Conv: 2-69 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-171 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-172 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-173 [1, 128, 80, 80] -- │ └─Conv: 2-70 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-174 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-175 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-176 [1, 128, 80, 80] -- │ └─Conv: 2-71 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-177 [1, 64, 80, 80] 73,795 │ │ └─BatchNorm2d: 3-178 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-179 [1, 64, 80, 80] -- │ └─Conv: 2-72 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-180 [1, 64, 80, 80] 36,931 │ │ └─BatchNorm2d: 3-181 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-182 [1, 64, 80, 80] -- │ └─Conv: 2-73 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-183 [1, 64, 80, 80] 36,931 │ │ └─BatchNorm2d: 3-184 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-185 [1, 64, 80, 80] -- │ └─Conv: 2-74 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-186 [1, 64, 80, 80] 36,931 │ │ └─BatchNorm2d: 3-187 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-188 [1, 64, 80, 80] -- │ └─Concat: 2-75 [1, 512, 80, 80] -- │ └─Conv: 2-76 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-189 [1, 128, 80, 80] 65,667 │ │ └─BatchNorm2d: 3-190 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-191 [1, 128, 80, 80] -- │ └─MP: 2-77 [1, 128, 40, 40] -- │ │ └─MaxPool2d: 3-192 [1, 128, 40, 40] -- │ └─Conv: 2-78 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-193 [1, 128, 40, 40] 16,515 │ │ └─BatchNorm2d: 3-194 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-195 [1, 128, 40, 40] -- │ └─Conv: 2-79 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-196 [1, 128, 80, 80] 16,515 │ │ └─BatchNorm2d: 3-197 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-198 [1, 128, 80, 80] -- │ └─Conv: 2-80 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-199 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-200 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-201 [1, 128, 40, 40] -- │ └─Concat: 2-81 [1, 512, 40, 40] -- │ └─Conv: 2-82 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-202 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-203 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-204 [1, 256, 40, 40] -- │ └─Conv: 2-83 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-205 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-206 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-207 [1, 256, 40, 40] -- │ └─Conv: 2-84 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-208 [1, 128, 40, 40] 295,043 │ │ └─BatchNorm2d: 3-209 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-210 [1, 128, 40, 40] -- │ └─Conv: 2-85 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-211 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-212 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-213 [1, 128, 40, 40] -- │ └─Conv: 2-86 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-214 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-215 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-216 [1, 128, 40, 40] -- │ └─Conv: 2-87 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-217 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-218 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-219 [1, 128, 40, 40] -- │ └─Concat: 2-88 [1, 1024, 40, 40] -- │ └─Conv: 2-89 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-220 [1, 256, 40, 40] 262,403 │ │ └─BatchNorm2d: 3-221 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-222 [1, 256, 40, 40] -- │ └─MP: 2-90 [1, 256, 20, 20] -- │ │ └─MaxPool2d: 3-223 [1, 256, 20, 20] -- │ └─Conv: 2-91 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-224 [1, 256, 20, 20] 65,795 │ │ └─BatchNorm2d: 3-225 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-226 [1, 256, 20, 20] -- │ └─Conv: 2-92 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-227 [1, 256, 40, 40] 65,795 │ │ └─BatchNorm2d: 3-228 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-229 [1, 256, 40, 40] -- │ └─Conv: 2-93 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-230 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-231 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-232 [1, 256, 20, 20] -- │ └─Concat: 2-94 [1, 1024, 20, 20] -- │ └─Conv: 2-95 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-233 [1, 512, 20, 20] 524,803 │ │ └─BatchNorm2d: 3-234 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-235 [1, 512, 20, 20] -- │ └─Conv: 2-96 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-236 [1, 512, 20, 20] 524,803 │ │ └─BatchNorm2d: 3-237 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-238 [1, 512, 20, 20] -- │ └─Conv: 2-97 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-239 [1, 256, 20, 20] 1,179,907 │ │ └─BatchNorm2d: 3-240 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-241 [1, 256, 20, 20] -- │ └─Conv: 2-98 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-242 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-243 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-244 [1, 256, 20, 20] -- │ └─Conv: 2-99 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-245 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-246 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-247 [1, 256, 20, 20] -- │ └─Conv: 2-100 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-248 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-249 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-250 [1, 256, 20, 20] -- │ └─Concat: 2-101 [1, 2048, 20, 20] -- │ └─Conv: 2-102 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-251 [1, 512, 20, 20] 1,049,091 │ │ └─BatchNorm2d: 3-252 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-253 [1, 512, 20, 20] -- │ └─RepConv: 2-103 [1, 256, 80, 80] -- │ │ └─Sequential: 3-254 [1, 256, 80, 80] 295,683 │ │ └─Sequential: 3-255 [1, 256, 80, 80] 33,539 │ │ └─SiLU: 3-256 [1, 256, 80, 80] -- │ └─RepConv: 2-104 [1, 512, 40, 40] -- │ │ └─Sequential: 3-257 [1, 512, 40, 40] 1,181,187 │ │ └─Sequential: 3-258 [1, 512, 40, 40] 132,611 │ │ └─SiLU: 3-259 [1, 512, 40, 40] -- │ └─RepConv: 2-105 [1, 1024, 20, 20] -- │ │ └─Sequential: 3-260 [1, 1024, 20, 20] 4,721,667 │ │ └─Sequential: 3-261 [1, 1024, 20, 20] 527,363 │ │ └─SiLU: 3-262 [1, 1024, 20, 20] -- │ └─IDetect: 2-106 [1, 25200, 6] -- ========================================================================================== Total params: 37,221,705 Trainable params: 37,221,420 Non-trainable params: 285 Total mult-adds (G): 52.29 ========================================================================================== Input size (MB): 4.92 Forward/backward pass size (MB): 1426.19 Params size (MB): 148.89 Estimated Total Size (MB): 1579.99 ========================================================================================== =============这里是使用load_state_dict =============The model exactly after :============ ===layer_name:=== ['model.0.conv.weight', 'model.0.conv.scale', 'model.0.conv.quantized_mode', 'model.0.bn.weight', 'model.0.bn.bias', 'model.0.bn.running_mean', 'model.0.bn.running_var', 'model.0.bn.num_batches_tracked', 'model.1.conv.weight', 'model.1.conv.scale', 'model.1.conv.quantized_mode', 'model.1.bn.weight', 'model.1.bn.bias', 'model.1.bn.running_mean', 'model.1.bn.running_var', 'model.1.bn.num_batches_tracked', 'model.2.conv.weight', 'model.2.conv.scale', 'model.2.conv.quantized_mode', 'model.2.bn.weight', 'model.2.bn.bias', 'model.2.bn.running_mean', 'model.2.bn.running_var', 'model.2.bn.num_batches_tracked', 'model.3.conv.weight', 'model.3.conv.scale', 'model.3.conv.quantized_mode', 'model.3.bn.weight', 'model.3.bn.bias', 'model.3.bn.running_mean', 'model.3.bn.running_var', 'model.3.bn.num_batches_tracked', 'model.4.conv.weight', 'model.4.conv.scale', 'model.4.conv.quantized_mode', 'model.4.bn.weight', 'model.4.bn.bias', 'model.4.bn.running_mean', 'model.4.bn.running_var', 'model.4.bn.num_batches_tracked', 'model.5.conv.weight', 'model.5.conv.scale', 'model.5.conv.quantized_mode', 'model.5.bn.weight', 'model.5.bn.bias', 'model.5.bn.running_mean', 'model.5.bn.running_var', 'model.5.bn.num_batches_tracked', 'model.6.conv.weight', 'model.6.conv.scale', 'model.6.conv.quantized_mode', 'model.6.bn.weight', 'model.6.bn.bias', 'model.6.bn.running_mean', 'model.6.bn.running_var', 'model.6.bn.num_batches_tracked', 'model.7.conv.weight', 'model.7.conv.scale', 'model.7.conv.quantized_mode', 'model.7.bn.weight', 'model.7.bn.bias', 'model.7.bn.running_mean', 'model.7.bn.running_var', 'model.7.bn.num_batches_tracked', 'model.8.conv.weight', 'model.8.conv.scale', 'model.8.conv.quantized_mode', 'model.8.bn.weight', 'model.8.bn.bias', 'model.8.bn.running_mean', 'model.8.bn.running_var', 'model.8.bn.num_batches_tracked', 'model.9.conv.weight', 'model.9.conv.scale', 'model.9.conv.quantized_mode', 'model.9.bn.weight', 'model.9.bn.bias', 'model.9.bn.running_mean', 'model.9.bn.running_var', 'model.9.bn.num_batches_tracked', 'model.11.conv.weight', 'model.11.conv.scale', 'model.11.conv.quantized_mode', 'model.11.bn.weight', 'model.11.bn.bias', 'model.11.bn.running_mean', 'model.11.bn.running_var', 'model.11.bn.num_batches_tracked', 'model.13.conv.weight', 'model.13.conv.scale', 'model.13.conv.quantized_mode', 'model.13.bn.weight', 'model.13.bn.bias', 'model.13.bn.running_mean', 'model.13.bn.running_var', 'model.13.bn.num_batches_tracked', 'model.14.conv.weight', 'model.14.conv.scale', 'model.14.conv.quantized_mode', 'model.14.bn.weight', 'model.14.bn.bias', 'model.14.bn.running_mean', 'model.14.bn.running_var', 'model.14.bn.num_batches_tracked', 'model.15.conv.weight', 'model.15.conv.scale', 'model.15.conv.quantized_mode', 'model.15.bn.weight', 'model.15.bn.bias', 'model.15.bn.running_mean', 'model.15.bn.running_var', 'model.15.bn.num_batches_tracked', 'model.17.conv.weight', 'model.17.conv.scale', 'model.17.conv.quantized_mode', 'model.17.bn.weight', 'model.17.bn.bias', 'model.17.bn.running_mean', 'model.17.bn.running_var', 'model.17.bn.num_batches_tracked', 'model.18.conv.weight', 'model.18.conv.scale', 'model.18.conv.quantized_mode', 'model.18.bn.weight', 'model.18.bn.bias', 'model.18.bn.running_mean', 'model.18.bn.running_var', 'model.18.bn.num_batches_tracked', 'model.19.conv.weight', 'model.19.conv.scale', 'model.19.conv.quantized_mode', 'model.19.bn.weight', 'model.19.bn.bias', 'model.19.bn.running_mean', 'model.19.bn.running_var', 'model.19.bn.num_batches_tracked', 'model.20.conv.weight', 'model.20.conv.scale', 'model.20.conv.quantized_mode', 'model.20.bn.weight', 'model.20.bn.bias', 'model.20.bn.running_mean', 'model.20.bn.running_var', 'model.20.bn.num_batches_tracked', 'model.21.conv.weight', 'model.21.conv.scale', 'model.21.conv.quantized_mode', 'model.21.bn.weight', 'model.21.bn.bias', 'model.21.bn.running_mean', 'model.21.bn.running_var', 'model.21.bn.num_batches_tracked', 'model.22.conv.weight', 'model.22.conv.scale', 'model.22.conv.quantized_mode', 'model.22.bn.weight', 'model.22.bn.bias', 'model.22.bn.running_mean', 'model.22.bn.running_var', 'model.22.bn.num_batches_tracked', 'model.24.conv.weight', 'model.24.conv.scale', 'model.24.conv.quantized_mode', 'model.24.bn.weight', 'model.24.bn.bias', 'model.24.bn.running_mean', 'model.24.bn.running_var', 'model.24.bn.num_batches_tracked', 'model.26.conv.weight', 'model.26.conv.scale', 'model.26.conv.quantized_mode', 'model.26.bn.weight', 'model.26.bn.bias', 'model.26.bn.running_mean', 'model.26.bn.running_var', 'model.26.bn.num_batches_tracked', 'model.27.conv.weight', 'model.27.conv.scale', 'model.27.conv.quantized_mode', 'model.27.bn.weight', 'model.27.bn.bias', 'model.27.bn.running_mean', 'model.27.bn.running_var', 'model.27.bn.num_batches_tracked', 'model.28.conv.weight', 'model.28.conv.scale', 'model.28.conv.quantized_mode', 'model.28.bn.weight', 'model.28.bn.bias', 'model.28.bn.running_mean', 'model.28.bn.running_var', 'model.28.bn.num_batches_tracked', 'model.30.conv.weight', 'model.30.conv.scale', 'model.30.conv.quantized_mode', 'model.30.bn.weight', 'model.30.bn.bias', 'model.30.bn.running_mean', 'model.30.bn.running_var', 'model.30.bn.num_batches_tracked', 'model.31.conv.weight', 'model.31.conv.scale', 'model.31.conv.quantized_mode', 'model.31.bn.weight', 'model.31.bn.bias', 'model.31.bn.running_mean', 'model.31.bn.running_var', 'model.31.bn.num_batches_tracked', 'model.32.conv.weight', 'model.32.conv.scale', 'model.32.conv.quantized_mode', 'model.32.bn.weight', 'model.32.bn.bias', 'model.32.bn.running_mean', 'model.32.bn.running_var', 'model.32.bn.num_batches_tracked', 'model.33.conv.weight', 'model.33.conv.scale', 'model.33.conv.quantized_mode', 'model.33.bn.weight', 'model.33.bn.bias', 'model.33.bn.running_mean', 'model.33.bn.running_var', 'model.33.bn.num_batches_tracked', 'model.34.conv.weight', 'model.34.conv.scale', 'model.34.conv.quantized_mode', 'model.34.bn.weight', 'model.34.bn.bias', 'model.34.bn.running_mean', 'model.34.bn.running_var', 'model.34.bn.num_batches_tracked', 'model.35.conv.weight', 'model.35.conv.scale', 'model.35.conv.quantized_mode', 'model.35.bn.weight', 'model.35.bn.bias', 'model.35.bn.running_mean', 'model.35.bn.running_var', 'model.35.bn.num_batches_tracked', 'model.37.conv.weight', 'model.37.conv.scale', 'model.37.conv.quantized_mode', 'model.37.bn.weight', 'model.37.bn.bias', 'model.37.bn.running_mean', 'model.37.bn.running_var', 'model.37.bn.num_batches_tracked', 'model.39.conv.weight', 'model.39.conv.scale', 'model.39.conv.quantized_mode', 'model.39.bn.weight', 'model.39.bn.bias', 'model.39.bn.running_mean', 'model.39.bn.running_var', 'model.39.bn.num_batches_tracked', 'model.40.conv.weight', 'model.40.conv.scale', 'model.40.conv.quantized_mode', 'model.40.bn.weight', 'model.40.bn.bias', 'model.40.bn.running_mean', 'model.40.bn.running_var', 'model.40.bn.num_batches_tracked', 'model.41.conv.weight', 'model.41.conv.scale', 'model.41.conv.quantized_mode', 'model.41.bn.weight', 'model.41.bn.bias', 'model.41.bn.running_mean', 'model.41.bn.running_var', 'model.41.bn.num_batches_tracked', 'model.43.conv.weight', 'model.43.conv.scale', 'model.43.conv.quantized_mode', 'model.43.bn.weight', 'model.43.bn.bias', 'model.43.bn.running_mean', 'model.43.bn.running_var', 'model.43.bn.num_batches_tracked', 'model.44.conv.weight', 'model.44.conv.scale', 'model.44.conv.quantized_mode', 'model.44.bn.weight', 'model.44.bn.bias', 'model.44.bn.running_mean', 'model.44.bn.running_var', 'model.44.bn.num_batches_tracked', 'model.45.conv.weight', 'model.45.conv.scale', 'model.45.conv.quantized_mode', 'model.45.bn.weight', 'model.45.bn.bias', 'model.45.bn.running_mean', 'model.45.bn.running_var', 'model.45.bn.num_batches_tracked', 'model.46.conv.weight', 'model.46.conv.scale', 'model.46.conv.quantized_mode', 'model.46.bn.weight', 'model.46.bn.bias', 'model.46.bn.running_mean', 'model.46.bn.running_var', 'model.46.bn.num_batches_tracked', 'model.47.conv.weight', 'model.47.conv.scale', 'model.47.conv.quantized_mode', 'model.47.bn.weight', 'model.47.bn.bias', 'model.47.bn.running_mean', 'model.47.bn.running_var', 'model.47.bn.num_batches_tracked', 'model.48.conv.weight', 'model.48.conv.scale', 'model.48.conv.quantized_mode', 'model.48.bn.weight', 'model.48.bn.bias', 'model.48.bn.running_mean', 'model.48.bn.running_var', 'model.48.bn.num_batches_tracked', 'model.50.conv.weight', 'model.50.conv.scale', 'model.50.conv.quantized_mode', 'model.50.bn.weight', 'model.50.bn.bias', 'model.50.bn.running_mean', 'model.50.bn.running_var', 'model.50.bn.num_batches_tracked', 'model.51.cv1.conv.weight', 'model.51.cv1.conv.scale', 'model.51.cv1.conv.quantized_mode', 'model.51.cv1.bn.weight', 'model.51.cv1.bn.bias', 'model.51.cv1.bn.running_mean', 'model.51.cv1.bn.running_var', 'model.51.cv1.bn.num_batches_tracked', 'model.51.cv2.conv.weight', 'model.51.cv2.conv.scale', 'model.51.cv2.conv.quantized_mode', 'model.51.cv2.bn.weight', 'model.51.cv2.bn.bias', 'model.51.cv2.bn.running_mean', 'model.51.cv2.bn.running_var', 'model.51.cv2.bn.num_batches_tracked', 'model.51.cv3.conv.weight', 'model.51.cv3.conv.scale', 'model.51.cv3.conv.quantized_mode', 'model.51.cv3.bn.weight', 'model.51.cv3.bn.bias', 'model.51.cv3.bn.running_mean', 'model.51.cv3.bn.running_var', 'model.51.cv3.bn.num_batches_tracked', 'model.51.cv4.conv.weight', 'model.51.cv4.conv.scale', 'model.51.cv4.conv.quantized_mode', 'model.51.cv4.bn.weight', 'model.51.cv4.bn.bias', 'model.51.cv4.bn.running_mean', 'model.51.cv4.bn.running_var', 'model.51.cv4.bn.num_batches_tracked', 'model.51.cv5.conv.weight', 'model.51.cv5.conv.scale', 'model.51.cv5.conv.quantized_mode', 'model.51.cv5.bn.weight', 'model.51.cv5.bn.bias', 'model.51.cv5.bn.running_mean', 'model.51.cv5.bn.running_var', 'model.51.cv5.bn.num_batches_tracked', 'model.51.cv6.conv.weight', 'model.51.cv6.conv.scale', 'model.51.cv6.conv.quantized_mode', 'model.51.cv6.bn.weight', 'model.51.cv6.bn.bias', 'model.51.cv6.bn.running_mean', 'model.51.cv6.bn.running_var', 'model.51.cv6.bn.num_batches_tracked', 'model.51.cv7.conv.weight', 'model.51.cv7.conv.scale', 'model.51.cv7.conv.quantized_mode', 'model.51.cv7.bn.weight', 'model.51.cv7.bn.bias', 'model.51.cv7.bn.running_mean', 'model.51.cv7.bn.running_var', 'model.51.cv7.bn.num_batches_tracked', 'model.52.conv.weight', 'model.52.conv.scale', 'model.52.conv.quantized_mode', 'model.52.bn.weight', 'model.52.bn.bias', 'model.52.bn.running_mean', 'model.52.bn.running_var', 'model.52.bn.num_batches_tracked', 'model.54.conv.weight', 'model.54.conv.scale', 'model.54.conv.quantized_mode', 'model.54.bn.weight', 'model.54.bn.bias', 'model.54.bn.running_mean', 'model.54.bn.running_var', 'model.54.bn.num_batches_tracked', 'model.56.conv.weight', 'model.56.conv.scale', 'model.56.conv.quantized_mode', 'model.56.bn.weight', 'model.56.bn.bias', 'model.56.bn.running_mean', 'model.56.bn.running_var', 'model.56.bn.num_batches_tracked', 'model.57.conv.weight', 'model.57.conv.scale', 'model.57.conv.quantized_mode', 'model.57.bn.weight', 'model.57.bn.bias', 'model.57.bn.running_mean', 'model.57.bn.running_var', 'model.57.bn.num_batches_tracked', 'model.58.conv.weight', 'model.58.conv.scale', 'model.58.conv.quantized_mode', 'model.58.bn.weight', 'model.58.bn.bias', 'model.58.bn.running_mean', 'model.58.bn.running_var', 'model.58.bn.num_batches_tracked', 'model.59.conv.weight', 'model.59.conv.scale', 'model.59.conv.quantized_mode', 'model.59.bn.weight', 'model.59.bn.bias', 'model.59.bn.running_mean', 'model.59.bn.running_var', 'model.59.bn.num_batches_tracked', 'model.60.conv.weight', 'model.60.conv.scale', 'model.60.conv.quantized_mode', 'model.60.bn.weight', 'model.60.bn.bias', 'model.60.bn.running_mean', 'model.60.bn.running_var', 'model.60.bn.num_batches_tracked', 'model.61.conv.weight', 'model.61.conv.scale', 'model.61.conv.quantized_mode', 'model.61.bn.weight', 'model.61.bn.bias', 'model.61.bn.running_mean', 'model.61.bn.running_var', 'model.61.bn.num_batches_tracked', 'model.63.conv.weight', 'model.63.conv.scale', 'model.63.conv.quantized_mode', 'model.63.bn.weight', 'model.63.bn.bias', 'model.63.bn.running_mean', 'model.63.bn.running_var', 'model.63.bn.num_batches_tracked', 'model.64.conv.weight', 'model.64.conv.scale', 'model.64.conv.quantized_mode', 'model.64.bn.weight', 'model.64.bn.bias', 'model.64.bn.running_mean', 'model.64.bn.running_var', 'model.64.bn.num_batches_tracked', 'model.66.conv.weight', 'model.66.conv.scale', 'model.66.conv.quantized_mode', 'model.66.bn.weight', 'model.66.bn.bias', 'model.66.bn.running_mean', 'model.66.bn.running_var', 'model.66.bn.num_batches_tracked', 'model.68.conv.weight', 'model.68.conv.scale', 'model.68.conv.quantized_mode', 'model.68.bn.weight', 'model.68.bn.bias', 'model.68.bn.running_mean', 'model.68.bn.running_var', 'model.68.bn.num_batches_tracked', 'model.69.conv.weight', 'model.69.conv.scale', 'model.69.conv.quantized_mode', 'model.69.bn.weight', 'model.69.bn.bias', 'model.69.bn.running_mean', 'model.69.bn.running_var', 'model.69.bn.num_batches_tracked', 'model.70.conv.weight', 'model.70.conv.scale', 'model.70.conv.quantized_mode', 'model.70.bn.weight', 'model.70.bn.bias', 'model.70.bn.running_mean', 'model.70.bn.running_var', 'model.70.bn.num_batches_tracked', 'model.71.conv.weight', 'model.71.conv.scale', 'model.71.conv.quantized_mode', 'model.71.bn.weight', 'model.71.bn.bias', 'model.71.bn.running_mean', 'model.71.bn.running_var', 'model.71.bn.num_batches_tracked', 'model.72.conv.weight', 'model.72.conv.scale', 'model.72.conv.quantized_mode', 'model.72.bn.weight', 'model.72.bn.bias', 'model.72.bn.running_mean', 'model.72.bn.running_var', 'model.72.bn.num_batches_tracked', 'model.73.conv.weight', 'model.73.conv.scale', 'model.73.conv.quantized_mode', 'model.73.bn.weight', 'model.73.bn.bias', 'model.73.bn.running_mean', 'model.73.bn.running_var', 'model.73.bn.num_batches_tracked', 'model.75.conv.weight', 'model.75.conv.scale', 'model.75.conv.quantized_mode', 'model.75.bn.weight', 'model.75.bn.bias', 'model.75.bn.running_mean', 'model.75.bn.running_var', 'model.75.bn.num_batches_tracked', 'model.77.conv.weight', 'model.77.conv.scale', 'model.77.conv.quantized_mode', 'model.77.bn.weight', 'model.77.bn.bias', 'model.77.bn.running_mean', 'model.77.bn.running_var', 'model.77.bn.num_batches_tracked', 'model.78.conv.weight', 'model.78.conv.scale', 'model.78.conv.quantized_mode', 'model.78.bn.weight', 'model.78.bn.bias', 'model.78.bn.running_mean', 'model.78.bn.running_var', 'model.78.bn.num_batches_tracked', 'model.79.conv.weight', 'model.79.conv.scale', 'model.79.conv.quantized_mode', 'model.79.bn.weight', 'model.79.bn.bias', 'model.79.bn.running_mean', 'model.79.bn.running_var', 'model.79.bn.num_batches_tracked', 'model.81.conv.weight', 'model.81.conv.scale', 'model.81.conv.quantized_mode', 'model.81.bn.weight', 'model.81.bn.bias', 'model.81.bn.running_mean', 'model.81.bn.running_var', 'model.81.bn.num_batches_tracked', 'model.82.conv.weight', 'model.82.conv.scale', 'model.82.conv.quantized_mode', 'model.82.bn.weight', 'model.82.bn.bias', 'model.82.bn.running_mean', 'model.82.bn.running_var', 'model.82.bn.num_batches_tracked', 'model.83.conv.weight', 'model.83.conv.scale', 'model.83.conv.quantized_mode', 'model.83.bn.weight', 'model.83.bn.bias', 'model.83.bn.running_mean', 'model.83.bn.running_var', 'model.83.bn.num_batches_tracked', 'model.84.conv.weight', 'model.84.conv.scale', 'model.84.conv.quantized_mode', 'model.84.bn.weight', 'model.84.bn.bias', 'model.84.bn.running_mean', 'model.84.bn.running_var', 'model.84.bn.num_batches_tracked', 'model.85.conv.weight', 'model.85.conv.scale', 'model.85.conv.quantized_mode', 'model.85.bn.weight', 'model.85.bn.bias', 'model.85.bn.running_mean', 'model.85.bn.running_var', 'model.85.bn.num_batches_tracked', 'model.86.conv.weight', 'model.86.conv.scale', 'model.86.conv.quantized_mode', 'model.86.bn.weight', 'model.86.bn.bias', 'model.86.bn.running_mean', 'model.86.bn.running_var', 'model.86.bn.num_batches_tracked', 'model.88.conv.weight', 'model.88.conv.scale', 'model.88.conv.quantized_mode', 'model.88.bn.weight', 'model.88.bn.bias', 'model.88.bn.running_mean', 'model.88.bn.running_var', 'model.88.bn.num_batches_tracked', 'model.90.conv.weight', 'model.90.conv.scale', 'model.90.conv.quantized_mode', 'model.90.bn.weight', 'model.90.bn.bias', 'model.90.bn.running_mean', 'model.90.bn.running_var', 'model.90.bn.num_batches_tracked', 'model.91.conv.weight', 'model.91.conv.scale', 'model.91.conv.quantized_mode', 'model.91.bn.weight', 'model.91.bn.bias', 'model.91.bn.running_mean', 'model.91.bn.running_var', 'model.91.bn.num_batches_tracked', 'model.92.conv.weight', 'model.92.conv.scale', 'model.92.conv.quantized_mode', 'model.92.bn.weight', 'model.92.bn.bias', 'model.92.bn.running_mean', 'model.92.bn.running_var', 'model.92.bn.num_batches_tracked', 'model.94.conv.weight', 'model.94.conv.scale', 'model.94.conv.quantized_mode', 'model.94.bn.weight', 'model.94.bn.bias', 'model.94.bn.running_mean', 'model.94.bn.running_var', 'model.94.bn.num_batches_tracked', 'model.95.conv.weight', 'model.95.conv.scale', 'model.95.conv.quantized_mode', 'model.95.bn.weight', 'model.95.bn.bias', 'model.95.bn.running_mean', 'model.95.bn.running_var', 'model.95.bn.num_batches_tracked', 'model.96.conv.weight', 'model.96.conv.scale', 'model.96.conv.quantized_mode', 'model.96.bn.weight', 'model.96.bn.bias', 'model.96.bn.running_mean', 'model.96.bn.running_var', 'model.96.bn.num_batches_tracked', 'model.97.conv.weight', 'model.97.conv.scale', 'model.97.conv.quantized_mode', 'model.97.bn.weight', 'model.97.bn.bias', 'model.97.bn.running_mean', 'model.97.bn.running_var', 'model.97.bn.num_batches_tracked', 'model.98.conv.weight', 'model.98.conv.scale', 'model.98.conv.quantized_mode', 'model.98.bn.weight', 'model.98.bn.bias', 'model.98.bn.running_mean', 'model.98.bn.running_var', 'model.98.bn.num_batches_tracked', 'model.99.conv.weight', 'model.99.conv.scale', 'model.99.conv.quantized_mode', 'model.99.bn.weight', 'model.99.bn.bias', 'model.99.bn.running_mean', 'model.99.bn.running_var', 'model.99.bn.num_batches_tracked', 'model.101.conv.weight', 'model.101.conv.scale', 'model.101.conv.quantized_mode', 'model.101.bn.weight', 'model.101.bn.bias', 'model.101.bn.running_mean', 'model.101.bn.running_var', 'model.101.bn.num_batches_tracked', 'model.102.rbr_dense.0.weight', 'model.102.rbr_dense.0.scale', 'model.102.rbr_dense.0.quantized_mode', 'model.102.rbr_dense.1.weight', 'model.102.rbr_dense.1.bias', 'model.102.rbr_dense.1.running_mean', 'model.102.rbr_dense.1.running_var', 'model.102.rbr_dense.1.num_batches_tracked', 'model.102.rbr_1x1.0.weight', 'model.102.rbr_1x1.0.scale', 'model.102.rbr_1x1.0.quantized_mode', 'model.102.rbr_1x1.1.weight', 'model.102.rbr_1x1.1.bias', 'model.102.rbr_1x1.1.running_mean', 'model.102.rbr_1x1.1.running_var', 'model.102.rbr_1x1.1.num_batches_tracked', 'model.103.rbr_dense.0.weight', 'model.103.rbr_dense.0.scale', 'model.103.rbr_dense.0.quantized_mode', 'model.103.rbr_dense.1.weight', 'model.103.rbr_dense.1.bias', 'model.103.rbr_dense.1.running_mean', 'model.103.rbr_dense.1.running_var', 'model.103.rbr_dense.1.num_batches_tracked', 'model.103.rbr_1x1.0.weight', 'model.103.rbr_1x1.0.scale', 'model.103.rbr_1x1.0.quantized_mode', 'model.103.rbr_1x1.1.weight', 'model.103.rbr_1x1.1.bias', 'model.103.rbr_1x1.1.running_mean', 'model.103.rbr_1x1.1.running_var', 'model.103.rbr_1x1.1.num_batches_tracked', 'model.104.rbr_dense.0.weight', 'model.104.rbr_dense.0.scale', 'model.104.rbr_dense.0.quantized_mode', 'model.104.rbr_dense.1.weight', 'model.104.rbr_dense.1.bias', 'model.104.rbr_dense.1.running_mean', 'model.104.rbr_dense.1.running_var', 'model.104.rbr_dense.1.num_batches_tracked', 'model.104.rbr_1x1.0.weight', 'model.104.rbr_1x1.0.scale', 'model.104.rbr_1x1.0.quantized_mode', 'model.104.rbr_1x1.1.weight', 'model.104.rbr_1x1.1.bias', 'model.104.rbr_1x1.1.running_mean', 'model.104.rbr_1x1.1.running_var', 'model.104.rbr_1x1.1.num_batches_tracked', 'model.105.anchors', 'model.105.anchor_grid', 'model.105.m.0.weight', 'model.105.m.0.bias', 'model.105.m.0.scale', 'model.105.m.0.quantized_mode', 'model.105.m.1.weight', 'model.105.m.1.bias', 'model.105.m.1.scale', 'model.105.m.1.quantized_mode', 'model.105.m.2.weight', 'model.105.m.2.bias', 'model.105.m.2.scale', 'model.105.m.2.quantized_mode', 'model.105.ia.0.implicit', 'model.105.ia.1.implicit', 'model.105.ia.2.implicit', 'model.105.im.0.implicit', 'model.105.im.1.implicit', 'model.105.im.2.implicit'] ======The whole model printing:====== Model( (model): Sequential( (0): Conv( (conv): MLUConv2d( 3, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([127.00000, 159.96063]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (1): Conv( (conv): MLUConv2d( 32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 1.88034, 242.17505]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (2): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 0.76717, 246.77040]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (3): Conv( (conv): MLUConv2d( 64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 2.16945, 659.72351]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (4): Conv( (conv): MLUConv2d( 128, 64, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.92369, 628.25122]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (5): Conv( (conv): MLUConv2d( 128, 64, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.92369, 475.93048]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (6): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 6.41040, 554.28021]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (7): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.03297, 458.72311]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (8): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.11026, 767.81110]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (9): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 9.95951, 859.82147]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (10): Concat() (11): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.41040, 273.06668]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (12): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (13): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 8.11695, 501.63162]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (14): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 8.11695, 685.81677]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (15): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 6.14557, 988.01898]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (16): Concat() (17): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 7.33305, 321.70190]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (18): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 7.33305, 435.67169]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (19): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 8.65670, 636.32050]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (20): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 9.67595, 407.03598]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (21): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 9.30384, 817.26947]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (22): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 8.85014, 893.03345]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (23): Concat() (24): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 5.68211, 132.76978]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (25): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (26): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.48813, 743.66260]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (27): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.48813, 1697.20068]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (28): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 6.09056, 2152.88989]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (29): Concat() (30): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 5.49991, 1761.86963]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (31): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 5.49991, 1102.68579]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (32): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 4.42657, 1578.73145]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (33): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.95019, 1872.87854]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (34): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.66529, 2285.30249]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (35): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.20015, 2676.22900]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (36): Concat() (37): Conv( (conv): MLUConv2d( 1024, 1024, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.37654, 1072.00830]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (38): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (39): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.97813, 1084.86340]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (40): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.97813, 1796.86353]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (41): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([2.79315e+00, 3.12897e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (42): Concat() (43): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.87262, 2285.30249]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (44): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.87262, 1101.51831]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (45): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.47857, 2038.96912]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (46): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.55337, 1933.79932]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (47): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 1.96922, 1937.40039]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (48): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.46558, 2126.48755]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (49): Concat() (50): Conv( (conv): MLUConv2d( 1024, 1024, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([1.96922e+00, 2.21594e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (51): SPPCSPC( (cv1): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.68248, 2660.82861]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([2.68248e+00, 2.89196e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.46504, 1826.83752]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv4): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.22379, 2116.75269]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): ModuleList( (0): MaxPool2d(kernel_size=5, stride=1, padding=2, dilation=1, ceil_mode=False) (1): MaxPool2d(kernel_size=9, stride=1, padding=4, dilation=1, ceil_mode=False) (2): MaxPool2d(kernel_size=13, stride=1, padding=6, dilation=1, ceil_mode=False) ) (cv5): Conv( (conv): MLUConv2d( 2048, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.67165, 1078.67700]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv6): Conv( (conv): MLUConv2d( 512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 6.95111, 1847.92896]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv7): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.59052, 1191.73425]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (52): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.05947, 1121.70776]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (53): Upsample(scale_factor=2.0, mode=nearest) (54): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.97813, 747.93964]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (55): Concat() (56): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.76734, 580.24762]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (57): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.76734, 463.21637]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (58): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.38502, 1523.25623]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (59): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.50332, 1491.58997]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (60): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.87678, 2361.82520]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (61): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.08313, 2051.02808]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (62): Concat() (63): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.09477, 108.37334]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (64): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.04031, 195.12077]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (65): Upsample(scale_factor=2.0, mode=nearest) (66): Conv( (conv): MLUConv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 6.48813, 98.59591]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (67): Concat() (68): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.61115, 108.46372]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (69): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.61115, 48.56161]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (70): Conv( (conv): MLUConv2d( 128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.58413, 58.05714]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (71): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 4.80973, 54.36789]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (72): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.31641, 159.56810]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (73): Conv( (conv): MLUConv2d( 64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 7.03452, 221.17007]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (74): Concat() (75): Conv( (conv): MLUConv2d( 512, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.58413, 81.63716]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (76): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (77): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 4.96912, 231.19644]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (78): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 4.96912, 157.72954]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (79): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([ 4.73306, 803.38531]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (80): Concat() (81): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.04031, 527.04358]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (82): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.04031, 605.22632]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (83): Conv( (conv): MLUConv2d( 256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.66400, 776.98584]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (84): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.54726, 1158.55676]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (85): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.66257, 892.26758]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (86): Conv( (conv): MLUConv2d( 128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.70854, 1283.63232]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (87): Concat() (88): Conv( (conv): MLUConv2d( 1024, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.63626, 660.56128]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (89): MP( (m): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) ) (90): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([1.42696e+00, 1.93921e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (91): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.42696, 1184.27319]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (92): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), scale=Parameter containing: tensor([2.26463e+00, 2.66936e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (93): Concat() (94): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([2.05947e+00, 3.00255e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (95): Conv( (conv): MLUConv2d( 1024, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.05947, 2010.40381]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (96): Conv( (conv): MLUConv2d( 512, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.60792, 2324.88037]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (97): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 2.54627, 2320.99048]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (98): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.66854, 1679.39307]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (99): Conv( (conv): MLUConv2d( 256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 3.67804, 2186.82910]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (100): Concat() (101): Conv( (conv): MLUConv2d( 2048, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 2.54117, 1505.62085]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (102): RepConv( (act): SiLU() (rbr_dense): Sequential( (0): MLUConv2d( 128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 4.96912, 691.28503]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) (rbr_1x1): Sequential( (0): MLUConv2d( 128, 256, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 4.96912, 801.52850]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) ) (103): RepConv( (act): SiLU() (rbr_dense): Sequential( (0): MLUConv2d( 256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([ 1.42696, 587.12415]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) (rbr_1x1): Sequential( (0): MLUConv2d( 256, 512, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.42696, 419.50967]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) ) (104): RepConv( (act): SiLU() (rbr_dense): Sequential( (0): MLUConv2d( 512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), scale=Parameter containing: tensor([1.79659e+00, 2.11460e+03]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) (rbr_1x1): Sequential( (0): MLUConv2d( 512, 1024, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.79659, 1612.99841]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): BatchNorm2d(1024, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) ) ) (105): IDetect( (m): ModuleList( (0): MLUConv2d( 256, 18, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 3.20508, 377.77197]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (1): MLUConv2d( 512, 18, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.24337, 245.83743]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) (2): MLUConv2d( 1024, 18, kernel_size=(1, 1), stride=(1, 1), scale=Parameter containing: tensor([ 1.27830, 1221.82507]), quantized_mode=Parameter containing: tensor([1], dtype=torch.int32), input_mean=None, input_std=None ) ) (ia): ModuleList( (0): ImplicitA() (1): ImplicitA() (2): ImplicitA() ) (im): ModuleList( (0): ImplicitM() (1): ImplicitM() (2): ImplicitM() ) ) ) ) ======The whole model torchinfo:====== Forward!!! ========================================================================================== Layer (type:depth-idx) Output Shape Param # ========================================================================================== Model -- -- ├─Sequential: 1 -- -- │ └─SPPCSPC: 2 -- -- │ │ └─ModuleList: 3-1 -- -- │ └─IDetect: 2 -- -- │ │ └─ModuleList: 3-2 -- 32,319 │ │ └─ModuleList: 3-3 -- 1,792 │ │ └─ModuleList: 3-4 -- 54 │ └─Conv: 2-1 [1, 32, 640, 640] -- │ │ └─MLUConv2d: 3-5 [1, 32, 640, 640] 899 │ │ └─BatchNorm2d: 3-6 [1, 32, 640, 640] 64 │ │ └─SiLU: 3-7 [1, 32, 640, 640] -- │ └─Conv: 2-2 [1, 64, 320, 320] -- │ │ └─MLUConv2d: 3-8 [1, 64, 320, 320] 18,499 │ │ └─BatchNorm2d: 3-9 [1, 64, 320, 320] 128 │ │ └─SiLU: 3-10 [1, 64, 320, 320] -- │ └─Conv: 2-3 [1, 64, 320, 320] -- │ │ └─MLUConv2d: 3-11 [1, 64, 320, 320] 36,931 │ │ └─BatchNorm2d: 3-12 [1, 64, 320, 320] 128 │ │ └─SiLU: 3-13 [1, 64, 320, 320] -- │ └─Conv: 2-4 [1, 128, 160, 160] -- │ │ └─MLUConv2d: 3-14 [1, 128, 160, 160] 73,859 │ │ └─BatchNorm2d: 3-15 [1, 128, 160, 160] 256 │ │ └─SiLU: 3-16 [1, 128, 160, 160] -- │ └─Conv: 2-5 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-17 [1, 64, 160, 160] 8,259 │ │ └─BatchNorm2d: 3-18 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-19 [1, 64, 160, 160] -- │ └─Conv: 2-6 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-20 [1, 64, 160, 160] 8,259 │ │ └─BatchNorm2d: 3-21 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-22 [1, 64, 160, 160] -- │ └─Conv: 2-7 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-23 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-24 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-25 [1, 64, 160, 160] -- │ └─Conv: 2-8 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-26 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-27 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-28 [1, 64, 160, 160] -- │ └─Conv: 2-9 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-29 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-30 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-31 [1, 64, 160, 160] -- │ └─Conv: 2-10 [1, 64, 160, 160] -- │ │ └─MLUConv2d: 3-32 [1, 64, 160, 160] 36,931 │ │ └─BatchNorm2d: 3-33 [1, 64, 160, 160] 128 │ │ └─SiLU: 3-34 [1, 64, 160, 160] -- │ └─Concat: 2-11 [1, 256, 160, 160] -- │ └─Conv: 2-12 [1, 256, 160, 160] -- │ │ └─MLUConv2d: 3-35 [1, 256, 160, 160] 65,795 │ │ └─BatchNorm2d: 3-36 [1, 256, 160, 160] 512 │ │ └─SiLU: 3-37 [1, 256, 160, 160] -- │ └─MP: 2-13 [1, 256, 80, 80] -- │ │ └─MaxPool2d: 3-38 [1, 256, 80, 80] -- │ └─Conv: 2-14 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-39 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-40 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-41 [1, 128, 80, 80] -- │ └─Conv: 2-15 [1, 128, 160, 160] -- │ │ └─MLUConv2d: 3-42 [1, 128, 160, 160] 32,899 │ │ └─BatchNorm2d: 3-43 [1, 128, 160, 160] 256 │ │ └─SiLU: 3-44 [1, 128, 160, 160] -- │ └─Conv: 2-16 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-45 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-46 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-47 [1, 128, 80, 80] -- │ └─Concat: 2-17 [1, 256, 80, 80] -- │ └─Conv: 2-18 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-48 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-49 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-50 [1, 128, 80, 80] -- │ └─Conv: 2-19 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-51 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-52 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-53 [1, 128, 80, 80] -- │ └─Conv: 2-20 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-54 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-55 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-56 [1, 128, 80, 80] -- │ └─Conv: 2-21 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-57 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-58 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-59 [1, 128, 80, 80] -- │ └─Conv: 2-22 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-60 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-61 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-62 [1, 128, 80, 80] -- │ └─Conv: 2-23 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-63 [1, 128, 80, 80] 147,587 │ │ └─BatchNorm2d: 3-64 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-65 [1, 128, 80, 80] -- │ └─Concat: 2-24 [1, 512, 80, 80] -- │ └─Conv: 2-25 [1, 512, 80, 80] -- │ │ └─MLUConv2d: 3-66 [1, 512, 80, 80] 262,659 │ │ └─BatchNorm2d: 3-67 [1, 512, 80, 80] 1,024 │ │ └─SiLU: 3-68 [1, 512, 80, 80] -- │ └─MP: 2-26 [1, 512, 40, 40] -- │ │ └─MaxPool2d: 3-69 [1, 512, 40, 40] -- │ └─Conv: 2-27 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-70 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-71 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-72 [1, 256, 40, 40] -- │ └─Conv: 2-28 [1, 256, 80, 80] -- │ │ └─MLUConv2d: 3-73 [1, 256, 80, 80] 131,331 │ │ └─BatchNorm2d: 3-74 [1, 256, 80, 80] 512 │ │ └─SiLU: 3-75 [1, 256, 80, 80] -- │ └─Conv: 2-29 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-76 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-77 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-78 [1, 256, 40, 40] -- │ └─Concat: 2-30 [1, 512, 40, 40] -- │ └─Conv: 2-31 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-79 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-80 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-81 [1, 256, 40, 40] -- │ └─Conv: 2-32 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-82 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-83 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-84 [1, 256, 40, 40] -- │ └─Conv: 2-33 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-85 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-86 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-87 [1, 256, 40, 40] -- │ └─Conv: 2-34 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-88 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-89 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-90 [1, 256, 40, 40] -- │ └─Conv: 2-35 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-91 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-92 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-93 [1, 256, 40, 40] -- │ └─Conv: 2-36 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-94 [1, 256, 40, 40] 590,083 │ │ └─BatchNorm2d: 3-95 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-96 [1, 256, 40, 40] -- │ └─Concat: 2-37 [1, 1024, 40, 40] -- │ └─Conv: 2-38 [1, 1024, 40, 40] -- │ │ └─MLUConv2d: 3-97 [1, 1024, 40, 40] 1,049,603 │ │ └─BatchNorm2d: 3-98 [1, 1024, 40, 40] 2,048 │ │ └─SiLU: 3-99 [1, 1024, 40, 40] -- │ └─MP: 2-39 [1, 1024, 20, 20] -- │ │ └─MaxPool2d: 3-100 [1, 1024, 20, 20] -- │ └─Conv: 2-40 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-101 [1, 512, 20, 20] 524,803 │ │ └─BatchNorm2d: 3-102 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-103 [1, 512, 20, 20] -- │ └─Conv: 2-41 [1, 512, 40, 40] -- │ │ └─MLUConv2d: 3-104 [1, 512, 40, 40] 524,803 │ │ └─BatchNorm2d: 3-105 [1, 512, 40, 40] 1,024 │ │ └─SiLU: 3-106 [1, 512, 40, 40] -- │ └─Conv: 2-42 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-107 [1, 512, 20, 20] 2,359,811 │ │ └─BatchNorm2d: 3-108 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-109 [1, 512, 20, 20] -- │ └─Concat: 2-43 [1, 1024, 20, 20] -- │ └─Conv: 2-44 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-110 [1, 256, 20, 20] 262,403 │ │ └─BatchNorm2d: 3-111 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-112 [1, 256, 20, 20] -- │ └─Conv: 2-45 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-113 [1, 256, 20, 20] 262,403 │ │ └─BatchNorm2d: 3-114 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-115 [1, 256, 20, 20] -- │ └─Conv: 2-46 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-116 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-117 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-118 [1, 256, 20, 20] -- │ └─Conv: 2-47 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-119 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-120 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-121 [1, 256, 20, 20] -- │ └─Conv: 2-48 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-122 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-123 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-124 [1, 256, 20, 20] -- │ └─Conv: 2-49 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-125 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-126 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-127 [1, 256, 20, 20] -- │ └─Concat: 2-50 [1, 1024, 20, 20] -- │ └─Conv: 2-51 [1, 1024, 20, 20] -- │ │ └─MLUConv2d: 3-128 [1, 1024, 20, 20] 1,049,603 │ │ └─BatchNorm2d: 3-129 [1, 1024, 20, 20] 2,048 │ │ └─SiLU: 3-130 [1, 1024, 20, 20] -- │ └─SPPCSPC: 2-52 [1, 512, 20, 20] -- │ │ └─Conv: 3-131 [1, 512, 20, 20] 525,827 │ │ └─Conv: 3-132 [1, 512, 20, 20] 2,360,835 │ │ └─Conv: 3-133 [1, 512, 20, 20] 263,683 │ │ └─Conv: 3-134 [1, 512, 20, 20] 1,050,115 │ │ └─Conv: 3-135 [1, 512, 20, 20] 2,360,835 │ │ └─Conv: 3-136 [1, 512, 20, 20] 525,827 │ │ └─Conv: 3-137 [1, 512, 20, 20] 525,827 │ └─Conv: 2-53 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-138 [1, 256, 20, 20] 131,331 │ │ └─BatchNorm2d: 3-139 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-140 [1, 256, 20, 20] -- │ └─Upsample: 2-54 [1, 256, 40, 40] -- │ └─Conv: 2-55 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-141 [1, 256, 40, 40] 262,403 │ │ └─BatchNorm2d: 3-142 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-143 [1, 256, 40, 40] -- │ └─Concat: 2-56 [1, 512, 40, 40] -- │ └─Conv: 2-57 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-144 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-145 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-146 [1, 256, 40, 40] -- │ └─Conv: 2-58 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-147 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-148 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-149 [1, 256, 40, 40] -- │ └─Conv: 2-59 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-150 [1, 128, 40, 40] 295,043 │ │ └─BatchNorm2d: 3-151 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-152 [1, 128, 40, 40] -- │ └─Conv: 2-60 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-153 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-154 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-155 [1, 128, 40, 40] -- │ └─Conv: 2-61 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-156 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-157 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-158 [1, 128, 40, 40] -- │ └─Conv: 2-62 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-159 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-160 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-161 [1, 128, 40, 40] -- │ └─Concat: 2-63 [1, 1024, 40, 40] -- │ └─Conv: 2-64 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-162 [1, 256, 40, 40] 262,403 │ │ └─BatchNorm2d: 3-163 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-164 [1, 256, 40, 40] -- │ └─Conv: 2-65 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-165 [1, 128, 40, 40] 32,899 │ │ └─BatchNorm2d: 3-166 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-167 [1, 128, 40, 40] -- │ └─Upsample: 2-66 [1, 128, 80, 80] -- │ └─Conv: 2-67 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-168 [1, 128, 80, 80] 65,667 │ │ └─BatchNorm2d: 3-169 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-170 [1, 128, 80, 80] -- │ └─Concat: 2-68 [1, 256, 80, 80] -- │ └─Conv: 2-69 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-171 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-172 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-173 [1, 128, 80, 80] -- │ └─Conv: 2-70 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-174 [1, 128, 80, 80] 32,899 │ │ └─BatchNorm2d: 3-175 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-176 [1, 128, 80, 80] -- │ └─Conv: 2-71 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-177 [1, 64, 80, 80] 73,795 │ │ └─BatchNorm2d: 3-178 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-179 [1, 64, 80, 80] -- │ └─Conv: 2-72 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-180 [1, 64, 80, 80] 36,931 │ │ └─BatchNorm2d: 3-181 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-182 [1, 64, 80, 80] -- │ └─Conv: 2-73 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-183 [1, 64, 80, 80] 36,931 │ │ └─BatchNorm2d: 3-184 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-185 [1, 64, 80, 80] -- │ └─Conv: 2-74 [1, 64, 80, 80] -- │ │ └─MLUConv2d: 3-186 [1, 64, 80, 80] 36,931 │ │ └─BatchNorm2d: 3-187 [1, 64, 80, 80] 128 │ │ └─SiLU: 3-188 [1, 64, 80, 80] -- │ └─Concat: 2-75 [1, 512, 80, 80] -- │ └─Conv: 2-76 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-189 [1, 128, 80, 80] 65,667 │ │ └─BatchNorm2d: 3-190 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-191 [1, 128, 80, 80] -- │ └─MP: 2-77 [1, 128, 40, 40] -- │ │ └─MaxPool2d: 3-192 [1, 128, 40, 40] -- │ └─Conv: 2-78 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-193 [1, 128, 40, 40] 16,515 │ │ └─BatchNorm2d: 3-194 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-195 [1, 128, 40, 40] -- │ └─Conv: 2-79 [1, 128, 80, 80] -- │ │ └─MLUConv2d: 3-196 [1, 128, 80, 80] 16,515 │ │ └─BatchNorm2d: 3-197 [1, 128, 80, 80] 256 │ │ └─SiLU: 3-198 [1, 128, 80, 80] -- │ └─Conv: 2-80 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-199 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-200 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-201 [1, 128, 40, 40] -- │ └─Concat: 2-81 [1, 512, 40, 40] -- │ └─Conv: 2-82 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-202 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-203 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-204 [1, 256, 40, 40] -- │ └─Conv: 2-83 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-205 [1, 256, 40, 40] 131,331 │ │ └─BatchNorm2d: 3-206 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-207 [1, 256, 40, 40] -- │ └─Conv: 2-84 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-208 [1, 128, 40, 40] 295,043 │ │ └─BatchNorm2d: 3-209 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-210 [1, 128, 40, 40] -- │ └─Conv: 2-85 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-211 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-212 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-213 [1, 128, 40, 40] -- │ └─Conv: 2-86 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-214 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-215 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-216 [1, 128, 40, 40] -- │ └─Conv: 2-87 [1, 128, 40, 40] -- │ │ └─MLUConv2d: 3-217 [1, 128, 40, 40] 147,587 │ │ └─BatchNorm2d: 3-218 [1, 128, 40, 40] 256 │ │ └─SiLU: 3-219 [1, 128, 40, 40] -- │ └─Concat: 2-88 [1, 1024, 40, 40] -- │ └─Conv: 2-89 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-220 [1, 256, 40, 40] 262,403 │ │ └─BatchNorm2d: 3-221 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-222 [1, 256, 40, 40] -- │ └─MP: 2-90 [1, 256, 20, 20] -- │ │ └─MaxPool2d: 3-223 [1, 256, 20, 20] -- │ └─Conv: 2-91 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-224 [1, 256, 20, 20] 65,795 │ │ └─BatchNorm2d: 3-225 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-226 [1, 256, 20, 20] -- │ └─Conv: 2-92 [1, 256, 40, 40] -- │ │ └─MLUConv2d: 3-227 [1, 256, 40, 40] 65,795 │ │ └─BatchNorm2d: 3-228 [1, 256, 40, 40] 512 │ │ └─SiLU: 3-229 [1, 256, 40, 40] -- │ └─Conv: 2-93 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-230 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-231 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-232 [1, 256, 20, 20] -- │ └─Concat: 2-94 [1, 1024, 20, 20] -- │ └─Conv: 2-95 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-233 [1, 512, 20, 20] 524,803 │ │ └─BatchNorm2d: 3-234 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-235 [1, 512, 20, 20] -- │ └─Conv: 2-96 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-236 [1, 512, 20, 20] 524,803 │ │ └─BatchNorm2d: 3-237 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-238 [1, 512, 20, 20] -- │ └─Conv: 2-97 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-239 [1, 256, 20, 20] 1,179,907 │ │ └─BatchNorm2d: 3-240 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-241 [1, 256, 20, 20] -- │ └─Conv: 2-98 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-242 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-243 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-244 [1, 256, 20, 20] -- │ └─Conv: 2-99 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-245 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-246 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-247 [1, 256, 20, 20] -- │ └─Conv: 2-100 [1, 256, 20, 20] -- │ │ └─MLUConv2d: 3-248 [1, 256, 20, 20] 590,083 │ │ └─BatchNorm2d: 3-249 [1, 256, 20, 20] 512 │ │ └─SiLU: 3-250 [1, 256, 20, 20] -- │ └─Concat: 2-101 [1, 2048, 20, 20] -- │ └─Conv: 2-102 [1, 512, 20, 20] -- │ │ └─MLUConv2d: 3-251 [1, 512, 20, 20] 1,049,091 │ │ └─BatchNorm2d: 3-252 [1, 512, 20, 20] 1,024 │ │ └─SiLU: 3-253 [1, 512, 20, 20] -- │ └─RepConv: 2-103 [1, 256, 80, 80] -- │ │ └─Sequential: 3-254 [1, 256, 80, 80] 295,683 │ │ └─Sequential: 3-255 [1, 256, 80, 80] 33,539 │ │ └─SiLU: 3-256 [1, 256, 80, 80] -- │ └─RepConv: 2-104 [1, 512, 40, 40] -- │ │ └─Sequential: 3-257 [1, 512, 40, 40] 1,181,187 │ │ └─Sequential: 3-258 [1, 512, 40, 40] 132,611 │ │ └─SiLU: 3-259 [1, 512, 40, 40] -- │ └─RepConv: 2-105 [1, 1024, 20, 20] -- │ │ └─Sequential: 3-260 [1, 1024, 20, 20] 4,721,667 │ │ └─Sequential: 3-261 [1, 1024, 20, 20] 527,363 │ │ └─SiLU: 3-262 [1, 1024, 20, 20] -- │ └─IDetect: 2-106 [1, 25200, 6] -- ========================================================================================== Total params: 37,221,705 Trainable params: 37,221,420 Non-trainable params: 285 Total mult-adds (G): 52.29 ========================================================================================== Input size (MB): 4.92 Forward/backward pass size (MB): 1426.19 Params size (MB): 148.89 Estimated Total Size (MB): 1579.99 ========================================================================================== Forward!!! CNRT: 4.10.1 a884a9a